Search

US-12619479-B2 - Artificial intelligence model management

US12619479B2US 12619479 B2US12619479 B2US 12619479B2US-12619479-B2

Abstract

A computer implemented method manages an artificial intelligence model. A number of processor units detect a change in a format used to exchange information between the artificial intelligence model and an application using the artificial intelligence model. The number of processor units changes the format of the information into an expected format used by the artificial intelligence model and the application. The number of processor units exchanges the information between the artificial intelligence model and the application using the expected format.

Inventors

  • Prateek Goyal
  • Manish Anand Bhide
  • Trent A. Gray-Donald

Assignees

  • INTERNATIONAL BUSINESS MACHINES CORPORATION

Dates

Publication Date
20260505
Application Date
20220104

Claims (17)

  1. 1 . A computer implemented method for managing an artificial intelligence model by an artificial intelligence manager running on a server data processing system coupled to a client device via a communication network, the computer implemented method comprising: receiving, by a number of processor units using the artificial intelligence manager running on the server data processing system, a request generated by an application using the artificial intelligence model, wherein the application is running on the client device; sending, by the number of processor units using the artificial intelligence manager running on the server data processing system, the request to the artificial intelligence model; receiving, by the number of processor units using the artificial intelligence manager running on the server data processing system, a reply generated by the artificial intelligence model; sending, by the number of processor units using the artificial intelligence manager running on the server data processing system, the reply to the application running on the client device; detecting, by the number of processor units using the artificial intelligence manager running on the server data processing system, a change in a data transmission format used to exchange information over the communication network between the artificial intelligence model and the application; changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format used to exchange the information over the communication network between the artificial intelligence model and the application into an expected data transmission format used by the artificial intelligence model and the application in response to the artificial intelligence manager detecting the change in the data transmission format; and exchanging, by the number of processor units using the artificial intelligence manager running on the server data processing system, the information between the artificial intelligence model and the application using the expected data transmission format in response to the artificial intelligence manager changing the data transmission format of the information, wherein the information exchanged between the artificial intelligence model and the application is selected from one of an output from the artificial intelligence model, an input to the artificial intelligence model, and the output from the artificial intelligence model and the input to the artificial intelligence model, wherein the input to the artificial intelligence model is a query request sent to the artificial intelligence model from the application via the communication network and the output from the artificial intelligence model is a query request result generated by the artificial intelligence model as a result of the artificial intelligence model receiving the query request sent to the artificial intelligence model from the application via the communication network.
  2. 2 . The computer implemented method of claim 1 , wherein detecting, by the number of processor units using the artificial intelligence manager running on the server data processing system, the change in the data transmission format used to exchange the information between the artificial intelligence model and the application comprises: detecting, by the number of processor units using the artificial intelligence manager running on the server data processing system, a framework change for the artificial intelligence model, wherein the framework change for the artificial intelligence model changes the data transmission format of how the information is exchanged between the artificial intelligence model and the application using the artificial intelligence model.
  3. 3 . The computer implemented method of claim 2 , wherein changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the information exchanged between the artificial intelligence model and the application comprises: changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the output of the artificial intelligence model from the framework format for the framework into a prior data transmission format used by the application; and changing, by the number of processor units using the artificial intelligence manager, the input to the artificial intelligence model from the prior data transmission format used by the application into the framework format for the artificial intelligence model.
  4. 4 . The computer implemented method of claim 1 , wherein detecting, by the number of processor units using the artificial intelligence manager running on the server data processing system, the change in the data transmission format used to exchange the information between the artificial intelligence model and the application comprises: profiling, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of an input and an output of the artificial intelligence model by comparing, by the number of processor units, the input and the output of the artificial intelligence model to a prior input and a prior output for the artificial intelligence model identified from a historical database of inputs and outputs for the artificial intelligence model to form a comparison; and using, by the number of processor units using the artificial intelligence manager running on the server data processing system, the comparison to detect the change in at least one of the input or the output of the artificial intelligence model.
  5. 5 . The computer implemented method of claim 1 , wherein changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the information comprises: automatically mapping, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the input from the application to the expected data transmission format used by the artificial intelligence model.
  6. 6 . The computer implemented method of claim 1 , wherein changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the information comprises: automatically mapping, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the output from the artificial intelligence model to the expected data transmission format used by the application.
  7. 7 . The computer implemented method of claim 1 , wherein changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the information comprises: receiving, by the number of processor units via the communication network using the artificial intelligence manager running on the server data processing system, a format mapping logic for mapping the data transmission format of the information into the expected data transmission format used by at least one of the artificial intelligence model or the application; and changing, by the number of processor units using the artificial intelligence manager running on the server data processing system, the data transmission format of the information into the expected data transmission format used by the artificial intelligence model and the application using the format mapping logic received by the number of processor units via the communication network.
  8. 8 . The computer implemented method of claim 1 , wherein the data transmission format of the information exchanged between the artificial intelligence model and the application is changed into the expected data transmission format used by the artificial intelligence model and the application using at least one of an application programming interface or code containing a logic for converting the information into the expected data transmission format.
  9. 9 . A computer system for managing an artificial intelligence model by an artificial intelligence model management system coupled to a client device via a communication network, comprising: a number of processor units, wherein the number of processor units executes program instructions to: receive, using the artificial intelligence model management system, a request generated by an application using the artificial intelligence model, wherein the application is running on the client device; send, using the artificial intelligence model management system, the request to the artificial intelligence model; receive, using the artificial intelligence model management system, a reply generated by the artificial intelligence model; send, using the artificial intelligence model management system, the reply to the application running on the client device; detect, using the artificial intelligence model management system, a change in a data transmission format used to exchange information over the communication network between an artificial intelligence model and the application; change, using the artificial intelligence model management system, the data transmission format used to exchange the information over the communication network between the artificial intelligence model and the application into the expected data transmission format used by the artificial intelligence model and the application in response to the artificial intelligence model management system detecting the change in the data transmission format; and exchange, using the artificial intelligence model management system, the information between the artificial intelligence model and the application using the expected data transmission format in response to the artificial intelligence model management system changing the data transmission format of the information, wherein the information exchanged between the artificial intelligence model and the application is selected from one of an output from the artificial intelligence model, an input to the artificial intelligence model, and the output from the artificial intelligence model and the input to the artificial intelligence model, wherein the input to the artificial intelligence model is a query request sent to the artificial intelligence model from the application via the communication network and the output from the artificial intelligence model is a query request result generated by the artificial intelligence model as a result of the artificial intelligence model receiving the query request sent to the artificial intelligence model from the application via the communication network.
  10. 10 . The computer system of claim 9 , wherein in detecting the change in the data transmission format used to exchange the information between the artificial intelligence model and the application, the number of processor units executes program instructions to: detect, using the artificial intelligence model management system, a framework change for the artificial intelligence model, wherein the framework change for the artificial intelligence model changes the data transmission format of how the information is exchanged between the artificial intelligence model and the application using the artificial intelligence model.
  11. 11 . The computer system of claim 9 , wherein in changing the data transmission format of the information exchanged between the artificial intelligence model and the application, the number of processor units executes program instructions to: changing, using the artificial intelligence model management system, the output of the artificial intelligence model from the framework format for the framework into a prior data transmission format used by the application; and changing, using the artificial intelligence model management system, the input to the artificial intelligence model from the prior data transmission format used by the application into the framework format for the artificial intelligence model.
  12. 12 . The computer system of claim 9 , wherein the computer system comprises a plurality of data processing systems coupled together via the communication network, wherein the application runs on a client data processing system of the plurality of data processing systems and the artificial intelligence model runs on a server data processing system of the plurality of data processing systems, and wherein in detecting the change in the data transmission format used to exchange the information between the artificial intelligence model and the application, the number of processor units executes program instructions to: profile, using the artificial intelligence model management system, the data transmission format of an input and an output of the artificial intelligence model, including compare the input and the output of the artificial intelligence model to a prior input and a prior output for the artificial intelligence model identified from a historical database of inputs and outputs for the artificial intelligence model to form a comparison; and use, using the artificial intelligence model management system, the comparison to detect the change in at least one of the input or the output of the artificial intelligence model.
  13. 13 . The computer system of claim 12 , wherein changing the data transmission format of the information, the number of processor units executes program instructions to: receive, via the communication network using the artificial intelligence model management system, a format mapping logic for mapping the data transmission format of the information into the expected data transmission format used by at least one of the artificial intelligence model or the application; and change, using the artificial intelligence model management system, the data transmission format of the information into the expected data transmission format used by the artificial intelligence model and the application using the format mapping logic received by the number of processor units via the communication network.
  14. 14 . The computer system of claim 9 , wherein in changing the data transmission format of the information, the number of processor units executes program instructions to: automatically map, using the artificial intelligence model management system, the data transmission format of the input from the application to the expected data transmission format used by the artificial intelligence model.
  15. 15 . The computer system of claim 9 , wherein in changing the data transmission format of the information, the number of processor units executes program instructions to: automatically map, using the artificial intelligence model management system, the data transmission format of the output from the artificial intelligence model to the expected data transmission format by used the application.
  16. 16 . The computer system of claim 9 , wherein the data transmission format of the information exchanged between the artificial intelligence model and the application is changed into the expected data transmission format used by the artificial intelligence model and the application using at least one of an application programming interface or code containing a logic for converting the information into the expected data transmission format.
  17. 17 . A computer program product for managing an artificial intelligence model by an artificial intelligence manager running on a server data processing system coupled to a client device via a communication network, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer system to cause the computer system to perform a method of: receiving, using the artificial intelligence manager running on the server data processing system, a request generated by an application using the artificial intelligence model, wherein the application is running on the client device; sending, using the artificial intelligence manager running on the server data processing system, the request to the artificial intelligence model; receiving, using the artificial intelligence manager running on the server data processing system, a reply generated by the artificial intelligence model; sending, using the artificial intelligence manager running on the server data processing system, the reply to the application running on the client device; detecting, using the artificial intelligence manager running on the server data processing system, a change in a data transmission format used to exchange information over the communication network between the artificial intelligence model and the application; changing, using the artificial intelligence manager running on the server data processing system, the data transmission format used to exchange the information over the communication network between the artificial intelligence model and the application into an expected data transmission format used by the artificial intelligence model and the application in response to the artificial intelligence manager detecting the change in the data transmission format; and exchanging, using the artificial intelligence manager running on the server data processing system, the information between the artificial intelligence model and the application using the expected data transmission format in response to the artificial intelligence manager changing the data transmission format of the information, wherein the information exchanged between the artificial intelligence model and the application is selected from one of an output from the artificial intelligence model, an input to the artificial intelligence model, and the output from the artificial intelligence model and the input to the artificial intelligence model, wherein the input to the artificial intelligence model is a query request sent to the artificial intelligence model from the application via the communication network and the output from the artificial intelligence model is a query request result generated by the artificial intelligence model as a result of the artificial intelligence model receiving the query request sent to the artificial intelligence model from the application via the communication network, and wherein detecting, by the number of processor units, the change in the data transmission format used to exchange information between the artificial intelligence model and an application comprises: profiling, using the artificial intelligence manager running on the server data processing system, the data transmission format of an input and an output of the artificial intelligence model by comparing, by the number of processor units, the input and the output of the artificial intelligence model to a prior input and a prior output for the artificial intelligence model identified from a historical database of inputs and outputs for the artificial intelligence model to form a comparison; and using, using the artificial intelligence manager running on the server data processing system, the comparison to detect the change in at least one of the input or the output of the artificial intelligence model.

Description

BACKGROUND 1. Field The disclosure relates generally to improved computer system and more specifically to managing artificial intelligence models using application programming interfaces. 2. Description of the Related Art Artificial intelligence models are models that have has intelligent behavior. For example, in artificial intelligence model can receive the environment and perform actions that maximize the probability of achieving a goal for which the artificial intelligence model has been constructed. An artificial model in some cases can mimic cognitive functions associated with the human mind. For example, an artificial intelligence model can include functions such as learning and problem-solving and can be based on the function of a human brain. An artificial intelligence model can change or evolve over time. As artificial intelligence models are used different issues can occur over time such as model drift, drop in accuracy, fairness issues, or other issues. When these issues are encountered, a new version of the artificial model can be created to address these issues. SUMMARY According to one illustrative embodiment, a computer implemented method manages an artificial intelligence model. A number of processor units detect a change in a format used to exchange information between the artificial intelligence model and an application using the artificial intelligence model. The number of processor units changes the format of the information into an expected format used by the artificial intelligence model and the application. The number of processor units exchanges the information between the artificial intelligence model and the application using the expected format. According to other illustrative embodiments, a computer system, and a computer program product for managing an artificial intelligence model are provided. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram illustrating a cloud computing environment in which illustrative embodiments may be implemented; FIG. 2 is a diagram illustrating abstraction model layers in accordance with an illustrative embodiment; FIG. 3 is a pictorial representation of a network of data processing systems in which illustrative embodiments may be implemented; FIG. 4 is a block diagram of an artificial intelligence environment in accordance with an illustrative embodiment; FIG. 5 is a dataflow diagram used in managing exchange of information between in artificial intelligence model and an application in accordance with an illustrative embodiment; FIG. 6 is an illustration of scalability for managing exchange of information between artificial intelligence models and applications in accordance with an illustrative embodiment; FIG. 7 is a flowchart of a process for managing an artificial intelligence model in accordance with an illustrative embodiment; FIG. 8 is a flowchart of a process for detecting a change in format in accordance with an illustrative embodiment; FIG. 9 is a flowchart of a process for detecting a change in format in accordance with an illustrative embodiment; FIG. 10 is a flowchart of a process for changing a format of information in accordance with an illustrative embodiment; FIG. 11 is a flowchart of a process for changing a format of information in accordance with an illustrative embodiment; FIG. 12 is a flowchart of a process for changing a format of information in accordance with an illustrative embodiment; FIG. 13 is a flowchart of a process for changing a format of information in accordance with an illustrative embodiment; and FIG. 14 is a block diagram of a data processing system in accordance with an illustrative embodiment. DETAILED DESCRIPTION The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention. The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded the